Three months ago, we announced the resurrection of FIBRE.

In that article, we argued that mining centralization remains one of Bitcoin’s most pressing tail risks. We outlined three vectors that entrench mining centralization within the Bitcoin network: pool centralization, supply chain concentration, and infrastructural asymmetry. We also described the emergence of “proxy pools,” where apparently independent pools source block templates from a single underlying entity, effectively centralizing template production in a cartel-like fashion.

FIBRE was our first public step toward addressing one part of that problem. Today, we are announcing the second step.

Localhost Research is launching a Mining Unit: a set of contributors working full time on reducing centralization pressures in the Bitcoin mining ecosystem. Alongside our continued efforts to support the FIBRE network, we are making a commitment to a keystone project: P2Poolv2.

We’ve selected this project as the primary focus area of our Mining Unit because it addresses two of the most fundamental drivers of pool centralization. First, it aims to break the dependency between variance smoothing and centralized pool structures. Second, P2Poolv2 enables payout structures to emerge competitively instead of being dictated by a pool operator.

Mining pools first emerged in the Bitcoin ecosystem in November 2010 with the launch of Slush Pool. Even then, participants understood that rising competition would make successful solo mining increasingly difficult. Slush Pool addressed this problem by allowing miners to coordinate their work in a cooperative arrangement, introducing what would soon become known as pooled mining.

Over the years, many pools have come and gone. And while block-discovery variance has remained a driving force behind the adoption of pools, the desire for smooth, predictable payouts has taken center stage. Interestingly, in the very same thread where Slush Pool announced itself, Satoshi made a prescient statement, describing a payout structure that would ultimately become known as FPPS, or Full Pay Per Share:

The instant gratification way would be to pay a fixed amount for each near-hit immediately, and the operator takes the risk from randomness of having more or less near-hits before a block is found.

In the last 10 years, FPPS — the payout structure that gives miners regular income regardless of pool luck — has emerged as a stabilizing but centralizing force in this evolving market dynamic. It is attractive for the same reason it is centralizing: it offloads payout-variance risk to a financier willing to backstop pool payouts. To do this effectively, significant capital is required. And it doesn’t always end well. We’ve seen pools collapse when their luck drops below the point at which they are able to backstop payments.

But this sort of collapse only reinforces the centralization risk. To reduce their variance exposure, the financier wants as much hashrate as possible. The result is a positive feedback loop: the larger the pool becomes, the less risky its financing becomes, and the more hashrate consolidates around it.

While the drive for payout-variance reduction is understandable, that guarantee has historically come bundled with outsourcing block template production entirely. Thankfully, the link between these two functions is not inextricable.

Stratum V2 (SV2) and other client-side block templating protocols have made this clear. They demonstrate that template construction can be performed by the miner, even while payouts continue to flow through a pool. This is a meaningful step, as it neutralizes one of the most acute centralization vectors while still enabling institutional miners to attach their hashrate to a pool that has regulatory and audit controls.

But there still remains a gap. While SV2 is a powerful primitive, supporting pools are still operated by a single entity and inherit all the associated risks. For this reason, there are miners who seek solutions that unbundle both block templating and payout management from pool operation. Enter P2Poolv2.

---

P2Poolv2 iterates on the original concept of P2Pool, an attempt to build an entirely decentralized pool structure. It failed for a number of reasons. However, none of these failings are fundamental limitations. With the right designs, a P2Pool-like protocol can once again rise. We believe P2Poolv2 reflects this possibility.

Alongside P2Poolv2’s lead maintainer, Jungly, we are committed to solidifying the underlying structures that ensure P2Poolv2 can scale and remain stable. We’ll work on bringing Compact Blocks-style relay to the P2Poolv2 node, simulating and improving the design of the sharechain, integrating P2Poolv2 into modern Mining Management Software (MMS) solutions, and more. ¹ Wherever possible, we want this work to land in a form that benefits the broader FOSS mining ecosystem, not just P2Poolv2 (e.g. we’d like to see SV2, DATUM, and other efforts converge around shared MMS infrastructure).

Infrastructure work aside, P2Poolv2 has to solve the payout problem, as coinbase payouts do not scale. In P2Poolv2, the solution to scaling payouts also happens to be the solution to miner payout variance reduction: make shares transferable. This is something we are very excited to work on.

In P2Poolv2, shares are not merely accounting entries inside a pool operator’s database. They are marketable claims. A miner who wants immediate, predictable revenue can sell shares for a fixed return in an open market. A market maker willing to warehouse variance can buy those shares in exchange for expected upside. FPPS-like products, PPS-like (Pay Per Share) products, and entirely new payout structures can emerge as competitive financial services on top of the share market, rather than as administrative policies enforced by a pool operator.

This payout model is one of the primary reasons we believe P2Poolv2 is a keystone project, and it represents one of many hard, interesting engineering problems still ahead of us.

Stay tuned for more information on our Mining Unit. We are excited to continue on our journey addressing this pressing issue in Bitcoin’s game mechanics.

A special thanks to Jungly for his leadership on P2Poolv2. We look forward to collaborating with him and the wider ecosystem for many years to come.